Research Finds That Electric Fields Help Neurons Fire 287
An anonymous reader writes "'[T]he brain is enveloped in countless overlapping electric fields, generated by the neural circuits of scores of communicating neurons. ... New work ... suggests that the fields do much more—and that they may, in fact, represent an additional form of neural communication. "In other words," says Anastassiou, the lead author of a paper about the work appearing in the journal Nature Neuroscience (abstract), "while active neurons give rise to extracellular fields, the same fields feed back to the neurons and alter their behavior," even though the neurons are not physically connected—a phenomenon known as ephaptic (or field) coupling. "So far, neural communication has been thought to occur almost entirely via traffic involving synapses, the junctions where one neuron connects to the next one. Our work suggests an additional means of neural communication through the extracellular space independent of synapses."' If this work is replicated, it could reveal that the brain is even more complicated and sophisticated than we thought — and raise new concerns about whether our cellphones and other electronic gizmos are affecting brain activity and memory. This is truly paradigm-busting work."
Re:It sounds like (Score:5, Interesting)
Or this could all be counted as interference that neurons though out all species have been fighting to over come, and hence make the job of coding AI easier relative to how the brains works.
Re:It sounds like (Score:5, Interesting)
This might push back the goalposts for the AI researchers.
AI != brain simulation. The stock markets run on AI. Cars and airplanes run on AI.
There is something known as the AI Effect [wikipedia.org] which tends to prevent us from recognizing applications of artificial intelligence as actual examples of AI, but, looking closely, you see that AI has little to do with the way the human brain works.
...In fact, that is kind of the "magic" of AI. It is an alien intelligence--at least to our way of thinking. So this discovery may be a major hurdle for those attempting to simulate or emulate a human brain, but the ever-progressing field of Artificial Intelligence cares little for such things.
Re:It sounds like (Score:5, Interesting)
If natural evolution wasn't able to perform a similar trick with the nervous system given around half a billion years to play with i'd be rather surprised.
I think just the opposite (Score:5, Interesting)
The brain is a noisy thing. Neural pathways are prone to error and so there are many for any given purpose processing the signals numerous times to ensure accuracy by aggregate measure. Low power devices find it difficult to maintain accurate signals and the brain is no exception. Signal redundancy and repetition would seem to be measures of compensation for the noisy environment that is the brain.
That electrical signals affect one another due to magnetic flux is nothing new. That the brain operates at low power and low signal requirements would seem to be factors that make it all seem possible in spite of all the noise that goes in on the brain.
I doubt seriously that the brain USES this type of signal processing and more likely that this is the type of thing that its redundancy systems are seeking to filter out. It also seems more likely to me that this is a source of hindrance to the brain rather than an enabler of its function. This could, however, serve to explain how seemingly disparate functions, senses and memories can be connected.
This is incredible. (Score:5, Interesting)
Telepathy? (Score:2, Interesting)
"Gizmos"? (Score:5, Interesting)
"...and raise new concerns about whether our cellphones and other electronic gizmos are affecting brain activity and memory."
Bullshit. What concerns where? That conclusion was not in the article. They didn't even talk about region-specific areas like memory.
I swear, people are so dedicated to perpetuating this stupid myth that consumer electronic devices interfere with our brains. Its been so thoroughly debunked that it's almost in the same realm as anti-vaccination/autism beliefs (except it doesn't get people killed.)
Not surprising (Score:5, Interesting)
I remember reading an article about a guy who was doing genetic algorithms with Xilinx chips, training them to recognize the words "stop" and "go" and set a line low or high accordingly. I can't find the article right now but I'll put in a better search later.
What he'd do is to say the word "stop" or "go" into a microphone and see what the circuit did. The genetic code was the array file input into the Xilinx chips, a string of binary data that his genetic routine would judge for fitness, splice, and retry.
He did several generations and eventually got a good working circuit. A series of ones and zeroes that recognized the words. It worked.
So he loaded the binary files into another board and it didn't work. Why? The genetic algorithm didn't view the circuits as digital. It was utilizing the gates as analog entities, each with it's unique characteristics to get the job done. When you move the code to another board it simply wouldn't work. There was more communication going on than the researcher's original notion imagined. He thought this was a binary exercise. Instead it turned out to be a subtle matter involving the shape of the response curves coming out of unique parts and electromagnetic field interaction. Nature didn't view this circuit as digital, it was more complex than that.
This article reminded me of that.
Shameless self publicity by the guest author (Score:2, Interesting)
"This is truly paradigm-busting work." In fact, probably not. I haven't read the background articles in detail, but the observed effect is either expected or completely unoriginal. It's not completely uninteresting, but to describe it as paradigm-busting is seriously overstating the article's significance.
In a little more detail. Using local extracellular current injections (not airy-fairy distant fields) to stimulate neurons is a technique going back more than a hundred years (Galvani...). It has also been known for decades that neurons themselves produce such currents. The only question is a quantitative one: are they big enough to matter for other neurons? The neuron-induced effect in this article is shown to be small but probably noticeable. I'm not sure however that even that is original (it's so paradigm-busting that I haven't yet taken the time to search out the older literature).
The paper is also amusing in that the author presumably pushing all of this publicity (Henry Markram) has only one listed contribution - helping to write the manuscript. Guidelines for responsible scientific communication at
http://www.sfn.org/index.aspx?pagename=responsibleConduct_authorsOfResearchManuscripts
indicate that mere writing is an insufficient contribution for authorship (note the "and"):
"1.6.1. SfN subscribes to the International Committee of Medical Journal Editors’ definition of authorship as being based on “1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published. Authors should meet conditions 1, 2, and 3..."
In other words, he did not contribute significantly to this work and is a guest author. Nature Neuroscience should have removed his name.
Re:I think just the opposite (Score:5, Interesting)
>>I doubt seriously that the brain USES this type of signal processing and more likely that this is the type of thing that its redundancy systems are seeking to filter out.
Don't make that claim unless you have evidence for it - you might be surprised.
In the neural circuits of crayfish, they actually work better with a certain amount of noise in the environment. It's a phenomenon known as stochastic resonance (http://en.wikipedia.org/wiki/Stochastic_resonance) which comes up in a lot of signal processing situations. I wouldn't be surprised if something similar was happening in our brains.